Advertisement

Editor's Choice – Severe Bleeding Risks of Direct Oral Anticoagulants in the Prevention and Treatment of Venous Thromboembolism: A Network Meta-Analysis of Randomised Controlled Trials

Published:December 29, 2021DOI:https://doi.org/10.1016/j.ejvs.2021.10.054

      Objective

      The aim of this study was to determine the severe bleeding safety of direct oral anticoagulants (DOACs) for the prevention and treatment of venous thromboembolism (VTE).

      Methods

      PubMed, EMBASE, Web of Science, and the Cochrane Library databases were searched up to 6 January 2021. The incidence of severe bleeding (major, gastrointestinal [GI], intracranial, and fatal) was investigated. Using frequentist network meta-analysis, interventions that were not compared directly could be compared indirectly by the 95% confidence interval (CI), making the search results more intuitive. Based on surface under the cumulative ranking curves (SUCRA), the relative ranking probability of each group was generated.

      Results

      Thirty-one randomised controlled trials (76 641 patients) were included. For the treatment of VTE, the risk of major bleeding with apixaban was significantly lower than dabigatran (odds ratio [OR] 2.10, 95% CI 1.07 – 4.12) and edoxaban (OR 2.64, 95% CI 1.36 – 5.15). The safety of the drugs was ranked from highest to lowest as follows: major bleeding: apixaban (SUCRA 98.0), rivaroxaban (SUCRA 69.6), dabigatran (SUCRA 50.7), edoxaban (SUCRA 26.5), and vitamin K antagonists (VKAs; SUCRA 5.1); GI bleeding: apixaban (SUCRA 80.7), rivaroxaban (SUCRA 66.8), edoxaban (SUCRA 62.3), VKAs (SUCRA 34.4), and dabigatran (SUCRA 5.8); intracranial bleeding: rivaroxaban (SUCRA 74.4), edoxaban (SUCRA 70.4), dabigatran (SUCRA 58.2), apixaban (SUCRA 44.4), and VKAs (SUCRA 5.6); fatal bleeding: edoxaban (SUCRA 82.7), rivaroxaban (SUCRA 59.2), dabigatran (SUCRA 48.6), apixaban (SUCRA 43.0), and VKAs (SUCRA 16.3). For the prevention of VTE, the risk of major bleeding with apixaban was significantly lower than rivaroxaban (OR 2.14, 95% CI 1.02 – 4.52). Among the four types of bleeding, apixaban had the lowest bleeding risk among DOACs (major bleeding: SUCRA 81.6; GI bleeding: SUCRA 75.4; intracranial bleeding: SUCRA 64.1; fatal bleeding: SUCRA 73.6).

      Conclusions

      For the treatment of VTE, in terms of major bleeding and GI bleeding, apixaban had the lowest bleeding risk; in terms of intracranial bleeding, rivaroxaban had the lowest bleeding risk; in terms of fatal bleeding, edoxaban had the lowest bleeding risk. For the prevention of VTE, apixaban had the lowest bleeding risk.

      Keywords

      Different direct oral anticoagulants (DOACs) used in the treatment and prevention of venous thromboembolism (VTE) have different bleeding risks. This systematic review and network meta-analysis analysed the risk of severe bleeding that occurred as a result of DOACs used for the prevention and treatment of VTE. For the treatment of VTE, in terms of major and gastrointestinal bleeding, apixaban had the lowest bleeding risk; for intracranial bleeding it was rivaroxaban, for fatal bleeding it was edoxaban, and for the prevention of VTE apixaban had the lowest bleeding risk.

      Introduction

      Venous thromboembolism (VTE) includes deep venous thrombosis (DVT) and pulmonary embolism (PE).
      • Kakkos S.K.
      • Gohel M.
      • Baekgaard N.
      • Bauersachs R.
      • Bellmunt-Montoya S.
      • Black S.A.
      • et al.
      Editor's Choice – European Society for Vascular Surgery (ESVS) 2021 Clinical Practice Guidelines on the Management of Venous Thrombosis.
      VTE has high mortality and high morbidity rates, and more than 900 000 people develop it each year. In the USA, there is an estimated annual incidence of one case per 120 people, which is similar to the incidence of stroke.
      • Heit J.A.
      • Spencer F.A.
      • White R.H.
      The epidemiology of venous thromboembolism.
      ,
      • Maynard G.
      Preventing Hospital-Associated Venous Thromboembolism: A Guide for Effective Quality Improvement.
      Therefore, the treatment and prevention of VTE is extremely important.
      Anticoagulant drugs mainly include heparin drugs, vitamin K antagonists (VKAs), and direct oral anticoagulants (DOACs), such as apixaban, dabigatran, edoxaban, and rivaroxaban. However, the disadvantages of some anticoagulant drugs (as represented by heparin drugs and VKAs) are becoming increasingly obvious. For example, parenteral administration of heparin drugs can cause local irritation, pain, and haematoma.
      • Shermock K.M.
      • Lau B.D.
      • Haut E.R.
      • Hobson D.B.
      • Ganetsky V.S.
      • Kraus P.S.
      • et al.
      Patterns of non-administration of ordered doses of venous thromboembolism prophylaxis: implications for novel intervention strategies.
      VKAs are oral anticoagulants, but they require laboratory monitoring of the international normalised ratio, and complex effects related to food–drug interactions have been noted.
      • Keeling D.
      • Baglin T.
      • Tait C.
      • Watson H.
      • Perry D.
      • Baglin C.
      • et al.
      Guidelines on oral anticoagulation with warfarin – fourth edition.
      However, DOACs do not require regular monitoring, there are fewer drug interactions, and the complexity of treatment management is lower, which is convenient for patients and clinicians.
      • Cohen A.T.
      • Berger S.E.
      • Milenković D.
      • Hill N.R.
      • Lister S.
      Anticoagulant selection for patients with VTE-Evidence from a systematic literature review of network meta-analyses.
      The U.S. Food and Drug Administration has approved five DOACs for the treatment or prevention of VTE, the latest of which is betrixaban.
      • Chaudhary R.
      • Sharma T.
      • Garg J.
      • Sukhi A.
      • Bliden K.
      • Tantry U.
      • et al.
      Direct oral anticoagulants: a review on the current role and scope of reversal agents.
      Although anticoagulants are effective in the prevention and treatment of VTE, the risk of bleeding events is increased vs. non-use of anticoagulants.
      • Harter K.
      • Levine M.
      • Henderson S.O.
      Anticoagulation drug therapy: a review.
      Fatal bleeding will directly lead to death, and other types of bleeding (major, intracranial, and gastrointestinal [GI]) also carry a high mortality rate. Compared with other anticoagulants, a meta-analysis reported that DOACs have higher efficacy in the prevention of VTE.
      • Gomez-Outes A.
      • Terleira-Fernandez A.I.
      • Suarez-Gea M.L.
      • Vargas-Castrillon E.
      Dabigatran, rivaroxaban, or apixaban versus enoxaparin for thromboprophylaxis after total hip or knee replacement: systematic review, meta-analysis, and indirect treatment comparisons.
      Nevertheless, conflicting results about whether DOACs have a lower risk of bleeding have been documented.
      • Büller H.R.
      • Prins M.H.
      • Lensin A.W.
      • Decousus H.
      • Jacobson B.F.
      • et al.
      EINSTEIN–PE Investigators
      Oral rivaroxaban for the treatment of symptomatic pulmonary embolism.
      ,
      • Cohen A.T.
      • Spiro T.E.
      • Büller H.R.
      • Haskell L.
      • Hu D.
      • Hull R.
      • et al.
      Rivaroxaban for thromboprophylaxis in acutely ill medical patients.
      A systematic review and network meta-analysis (NMA) of the four types of severe bleeding related to DOACs used for the prevention or treatment of VTE was conducted, in order to provide references for clinical application.

      Methods

      Review design

      The Preferred reporting items for Systematic Reviews and Meta-Analysis (PRISMA)-NMA extension guideline and checklist for reporting systematic reviews and NMA were followed (Supplementary Table S1).
      • Hutton B.
      • Salanti G.
      • Caldwell D.M.
      • Chaimani A.
      • Schmid C.H.
      • Cameron C.
      • et al.
      The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations.

      Criteria for considering studies

      The inclusion criteria were (i) randomised controlled trials (RCTs); (ii) comparison between DOACs commonly used in clinical practice (apixaban, edoxaban, dabigatran, and rivaroxaban) and control, including low molecular weight heparins (LMWHs) and VKAs; (iii) patients aged ≥ 18 years undergoing prevention or treatment of VTE; (iv) outcome indicators, including at least one incident of major, GI, intracranial, or fatal bleeding; and (v) for RCTs comparing various doses of DOACs with a control dose, only the regular doses (see Appendix S1) were included in this study.
      The exclusion criteria were (i) cancer associated thrombosis; (ii) combined use of other antithrombotic drugs; and (iii) repeated studies, incomplete/unavailable original research data.

      Types of outcome measures

      Safety results included four types of bleeding (major, GI, intracranial, and fatal). The definition of major bleeding was consistent across the all trials and the major bleeding was defined according to International Society on Thrombosis and Haemostasis (ISTH; Appendix S1).
      • Schulman S.
      • Kearon C.
      Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients.

      Search methods for the identification of studies

      A systematic search was conducted in PubMed, EMBASE, Web of Science, and the Cochrane Library databases for relevant articles published in English before 6 January 2021. The key words and retrieval strategies are presented in Appendix S1 and Supplementary Table S2. For completeness, the reference lists of relevant studies found in database searches were checked. The literature searches and screening were carried out independently by two researchers (J.C. and M.L.).

      Selection of studies and data management

      Two researchers (J.C. and M.L.) independently extracted data from relevant RCTs. These data included first author and publication year; number of patients; age; sex ratio; number of patients with bleeding (major, GI, intracranial, and fatal); and duration of follow up. Disputes were discussed and resolved by a third researcher (S.W.).

      Assessment of risk of bias of included studies and GRADE assessment

      Included studies were assessed using the Risk of Bias 2 tool (RoB 2);
      • Sterne J.A.C.
      • Savović J.
      • Page M.J.
      • Elbers R.G.
      • Blencowe N.S.
      • Boutron I.
      • et al.
      RoB 2: a revised tool for assessing risk of bias in randomised trials.
      the risk of a bias in each RCT was evaluated by two independent researchers (J.C. and M.L.). During evaluation, disputes were resolved by the evaluation of another researcher (S.W.). The quality of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, which classified the quality of evidence as “high”, “moderate”, “low”, or “very low”.
      • Salanti G.
      • Del Giovane C.
      • Chaimani A.
      • Caldwell D.M.
      • Higgins J.P.
      Evaluating the quality of evidence from a network meta-analysis.

      Statistical analyses

      The bleeding that occurred as a result of anticoagulants used to treat and prevent VTE was analysed separately, and the safety of the drugs for the four types of bleeding was ranked. In the present study, Stata (StataCorp, College Station, TX, USA) based on a frequentist framework was used for the NMA. First, the extracted data were sorted and paired, and then imported into Stata 14.0. The node size and thickness of the line in the network evidence map represent the number of patients included in the corresponding intervention and number of directly compared interventions. If the funnel chart was roughly symmetrical and the Egger test p > .05, there was no publication bias in the NMA. The odds ratio (OR) and 95% confidence interval (CI) from forest plots were used to evaluate whether the difference in the incidence of bleeding was significant between the interventions. If the CI did not cross the value 1, then the result was considered significant and there was a significant difference between the two. The surface under the cumulative ranking curves (SUCRA) was used to assess the bleeding risk. The larger the area under SUCRA, the lower the risk of bleeding and the higher the safety. To examine transitivity, the population, intervention, comparison, and outcome (PICO) methodology was used to assess each of the selected articles. Sensitivity analysis was performed for indications, and studies of knee or hip arthroplasty and acute VTE were selected separately in the prevention and treatment studies. The rank probability was calculated again. If there was no significant change, the outcome of the NMA was considered to be reliable.

      Results

      Study selection and study characteristics

      The search strategy identified 9 076 studies, of which 31 articles were included for analyses (Fig. 1).
      • Büller H.R.
      • Prins M.H.
      • Lensin A.W.
      • Decousus H.
      • Jacobson B.F.
      • et al.
      EINSTEIN–PE Investigators
      Oral rivaroxaban for the treatment of symptomatic pulmonary embolism.
      ,
      • Cohen A.T.
      • Spiro T.E.
      • Büller H.R.
      • Haskell L.
      • Hu D.
      • Hull R.
      • et al.
      Rivaroxaban for thromboprophylaxis in acutely ill medical patients.
      ,
      • Agnelli G.
      • Buller H.R.
      • Cohen A.
      • Curto M.
      • Gallus A.S.
      • Johnson M.
      • et al.
      Oral apixaban for the treatment of acute venous thromboembolism.
      • Buller H.
      • Deitchman D.
      • Prins M.
      • Segers A.
      Botticelli Investigators, Writing Committe
      Efficacy and safety of the oral direct factor Xa inhibitor apixaban for symptomatic deep vein thrombosis. The Botticelli DVT dose-ranging study.
      • Buller H.R.
      • Lensing A.W.
      • Prins M.H.
      • Agnelli G.
      • Cohen A.
      • Gallus A.S.
      • et al.
      A dose-ranging study evaluating once-daily oral administration of the factor Xa inhibitor rivaroxaban in the treatment of patients with acute symptomatic deep vein thrombosis: the Einstein-DVT Dose-Ranging Study.
      • Büller H.R.
      • Décousus H.
      • Grosso M.A.
      • Mercuri M.
      • Middeldorp S.
      • et al.
      Hokusai-VTE Investigators
      Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism.
      • Bauersachs R.
      • Berkowitz S.D.
      • Brenner B.
      • Buller H.R.
      • Decousus H.
      • et al.
      EINSTEIN Investigators
      Oral rivaroxaban for symptomatic venous thromboembolism.
      • Eriksson B.I.
      • Dahl O.E.
      • Büller H.R.
      • Hettiarachchi R.
      • Rosencher N.
      • Bravo M.L.
      • et al.
      A new oral direct thrombin inhibitor, dabigatran etexilate, compared with enoxaparin for prevention of thromboembolic events following total hip or knee replacement: the BISTRO II randomized trial.
      • Eriksson B.I.
      • Borris L.C.
      • Dahl O.E.
      • Haas S.
      • Huisman M.V.
      • Kakkar A.K.
      • et al.
      A once-daily, oral, direct Factor Xa inhibitor, rivaroxaban (BAY 59-7939), for thromboprophylaxis after total hip replacement.
      • Eriksson B.I.
      • Dahl O.E.
      • Rosencher N.
      • Kurth A.A.
      • van Dijk C.N.
      • Frostick S.P.
      • et al.
      Dabigatran etexilate versus enoxaparin for prevention of venous thromboembolism after total hip replacement: a randomised, double-blind, non-inferiority trial.
      • Eriksson B.I.
      • Dahl O.E.
      • Rosencher N.
      • Kurth A.A.
      • van Dijk C.N.
      • Frostick S.P.
      • et al.
      Oral dabigatran etexilate vs. subcutaneous enoxaparin for the prevention of venous thromboembolism after total knee replacement: the RE-MODEL randomized trial.
      • Eriksson B.I.
      • Borris L.C.
      • Friedman R.J.
      • Haas S.
      • Huisman M.V.
      • Kakkar A.K.
      • et al.
      Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty.
      • Eriksson B.I.
      • Dahl O.E.
      • Huo M.H.
      • Kurth A.A.
      • Hantel S.
      • Hermansson K.
      • et al.
      Oral dabigatran versus enoxaparin for thromboprophylaxis after primary total hip arthroplasty (RE-NOVATE II∗). A randomised, double-blind, non-inferiority trial.
      • Fuji T.
      • Wang C.J.
      • Fujita S.
      • Kawai Y.
      • Kimura T.
      • Tachibana S.
      Safety and efficacy of edoxaban, an oral factor xa inhibitor, for thromboprophylaxis after total hip arthroplasty in Japan and Taiwan.
      • Fuji T.
      • Wang C.J.
      • Fujita S.
      • Kawai Y.
      • Nakamura M.
      • Kimura T.
      • et al.
      Safety and efficacy of edoxaban, an oral factor Xa inhibitor, versus enoxaparin for thromboprophylaxis after total knee arthroplasty: the STARS E-3 trial.
      • Fuji T.
      • Fujita S.
      • Kawai Y.
      • Nakamura M.
      • Kimura T.
      • Kiuchi Y.
      • et al.
      Safety and efficacy of edoxaban in patients undergoing hip fracture surgery.
      • Fuji T.
      • Fujita S.
      • Kawai Y.
      • Nakamura M.
      • Kimura T.
      • Fukuzawa M.
      • et al.
      Efficacy and safety of edoxaban versus enoxaparin for the prevention of venous thromboembolism following total hip arthroplasty: STARS J-V.
      • Ginsberg J.S.
      • Davidson B.L.
      • Comp P.C.
      • Francis C.W.
      • Friedman R.J.
      • et al.
      RE-MOBILIZE Writing Committee
      Oral thrombin inhibitor dabigatran etexilate vs North American enoxaparin regimen for prevention of venous thromboembolism after knee arthroplasty surgery.
      • Kakkar A.K.
      • Brenner B.
      • Dahl O.E.
      • Eriksson B.I.
      • Mouret P.
      • Muntz J.
      • et al.
      Extended duration rivaroxaban versus short-term enoxaparin for the prevention of venous thromboembolism after total hip arthroplasty: a double-blind, randomised controlled trial.
      • Lassen M.R.
      • Davidson B.L.
      • Gallus A.
      • Pineo G.
      • Ansell J.
      • Deitchman D.
      The efficacy and safety of apixaban, an oral, direct factor Xa inhibitor, as thromboprophylaxis in patients following total knee replacement.
      • Lassen M.R.
      • Ageno W.
      • Borris L.C.
      • Lieberman J.R.
      • Rosencher N.
      • Bandel T.J.
      • et al.
      Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty.
      • Lassen M.R.
      • Raskob G.E.
      • Gallus A.
      • Pineo G.
      • Chen D.
      • Portman R.J.
      Apixaban or enoxaparin for thromboprophylaxis after knee replacement.
      • Lassen M.R.
      • Raskob G.E.
      • Gallus A.
      • Pineo G.
      • Chen D.
      • Hornick P.
      • et al.
      Apixaban versus enoxaparin for thromboprophylaxis after knee replacement (ADVANCE-2): a randomised double-blind trial.
      • Lassen M.R.
      • Raskob G.E.
      • Gallus A.
      • Pineo G.
      • Chen D.
      • Hornick P.
      • et al.
      Apixaban versus enoxaparin for thromboprophylaxis after hip replacement.
      • Nakamura M.
      • Nishikawa M.
      • Komuro I.
      • Kitajima I.
      • Uetsuka Y.
      • Yamagami T.
      • et al.
      Apixaban for the treatment of japanese subjects with acute venous thromboembolism (AMPLIFY-J Study).
      • Raskob G.
      • Cohen A.T.
      • Eriksson B.I.
      • Puskas D.
      • Shi M.
      • Bocanegra T.
      • et al.
      Oral direct factor Xa inhibition with edoxaban for thromboprophylaxis after elective total hip replacement. A randomised double-blind dose-response study.
      • Schulman S.
      • Kearon C.
      • Kakkar A.K.
      • Mismetti P.
      • Schellong S.
      • Eriksson H.
      • et al.
      Dabigatran versus warfarin in the treatment of acute venous thromboembolism.
      • Schulman S.
      • Kakkar A.K.
      • Goldhaber S.Z.
      • Schellong S.
      • Eriksson H.
      • Mismetti P.
      • et al.
      Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.
      • Schulman S.
      • Kearon C.
      • Kakkar A.K.
      • Schellong S.
      • Eriksson H.
      • Baanstra D.
      • et al.
      Extended use of dabigatran, warfarin, or placebo in venous thromboembolism.
      • Samama C.M.
      • Laporte S.
      • Rosencher N.
      • Girard P.
      • Llau J.
      • Mouret P.
      • et al.
      Rivaroxaban or enoxaparin in nonmajor orthopedic surgery.
      • Turpie A.G.
      • Lassen M.R.
      • Davidson B.L.
      • Bauer K.A.
      • Gent M.
      • Kwong L.M.
      • et al.
      Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty (RECORD4): a randomised trial.
      The 31 RCTs involved 76 641 patients. Ten articles discussed the treatment of VTE and 21 the prevention of VTE. The characteristics of the 31 included RCTs are shown in Supplementary Table S3.
      Figure 1
      Figure 1Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) flow diagram showing the selection of randomised controlled trials (RCT) on the risk of severe bleeding related to anticoagulants for the prevention and treatment of venous thromboembolism (VTE).

      Presentation of network geometry and structure

      The network evidence diagram between interventions is provided in Supplementary Figure S1. The size of the node represents the sample size of the intervention and the thickness of the line represents the number of studies between interventions. For treatment of VTE, the test group included four DOACs and the control group VKAs; for the prevention of VTE, the test group included four DOACs and the control group LMWH. In both the treatment and prevention of VTE, DOACs were indirectly compared with each other. Comparisons between interventions are shown in the Supplementary Table S4.

      Risk of bias in included studies and GRADE evaluation results

      The results of the bias risk assessment and the reasons for the judgement of the included articles are given in Supplementary Tables S5 and S6; the quality of the included RCTs was good and the risk of a publication bias was low. Four articles were not double blind trials and were assessed to be at “high risk” of bias.
      • Büller H.R.
      • Prins M.H.
      • Lensin A.W.
      • Decousus H.
      • Jacobson B.F.
      • et al.
      EINSTEIN–PE Investigators
      Oral rivaroxaban for the treatment of symptomatic pulmonary embolism.
      ,
      • Bauersachs R.
      • Berkowitz S.D.
      • Brenner B.
      • Buller H.R.
      • Decousus H.
      • et al.
      EINSTEIN Investigators
      Oral rivaroxaban for symptomatic venous thromboembolism.
      ,
      • Fuji T.
      • Fujita S.
      • Kawai Y.
      • Nakamura M.
      • Kimura T.
      • Kiuchi Y.
      • et al.
      Safety and efficacy of edoxaban in patients undergoing hip fracture surgery.
      ,
      • Nakamura M.
      • Nishikawa M.
      • Komuro I.
      • Kitajima I.
      • Uetsuka Y.
      • Yamagami T.
      • et al.
      Apixaban for the treatment of japanese subjects with acute venous thromboembolism (AMPLIFY-J Study).
      The GRADE evaluation results are given in Supplementary Tables S7 and S8. The evaluation results showed that most of the studies provided moderate evidence and the small studies provided low and high evidence, indicating that the overall level of evidence was moderate.

      Assessment of inconsistency and transitivity assumption

      All evidence was obtained from direct comparisons, and inconsistency could not be evaluated because the network did not form a closed loop. In the transitivity assumption, indicators such as sex ratio, age, dose, route of administration, outcome definition, follow up time, and duration of treatment were assessed. The studies selected in the network meta-analysis differed with regard to length of follow up and duration of treatment, but the remaining PICO parameters were basically uniform, thus satisfying the transitivity assumption.

      Network meta-analysis results for the treatment of venous thromboembolism

      Major bleeding

      NMA results for major bleeding are shown in Figure 2A. Compared with apixaban, dabigatran (OR 2.10, 95% CI 1.07 – 4.12), edoxaban (OR 2.64, 95% CI 1.36 – 5.15), and VKAs (OR 3.12; 95% CI 1.78 – 5.47) had a higher risk of bleeding, and the difference was significant. Compared with VKAs, dabigatran (OR 1.48, 95% CI 1.02 – 2.15) and rivaroxaban (OR 1.82, 95% CI 1.24 – 2.68) had a lower risk of bleeding, and the difference was significant.
      Figure 2
      Figure 2Forest plots of (A) major bleeding, (B) gastrointestinal bleeding, (C) intracranial bleeding, and (D) fatal bleeding related to the treatment of venous thromboembolism (VTE). Asterisk indicates statistical significance. CI = confidence interval; VKA = vitamin K antagonist; Prl=prediction interval.
      In the ranking of the cumulative probability of major bleeding (Table 1 and Supplementary Fig. S2A), apixaban (SUCRA 98.0) had the highest safety, followed by rivaroxaban (SUCRA 69.6), dabigatran (SUCRA 50.7), edoxaban (SUCRA 26.5), and VKAs (SUCRA 5.1).
      Table 1Bleeding risk ranking of anticoagulants for major bleeding (MB), gastrointestinal bleeding (GIB), intracranial haemorrhage (ICH), and fatal bleeding (FB) in the prevention and treatment of venous thromboembolism presented as surface under the cumulative ranking curve (SUCRA) values, where the higher the ranking, the lower the risk of bleeding
      RankMBGIBICHFB
      DrugSUCRA valueDrugSUCRA valueDrugSUCRA valueDrugSUCRA value
      Treatment
      1Apixaban98.0Apixaban80.7Rivaroxaban74.4Edoxaban82.7
      2Rivaroxaban69.6Rivaroxaban66.8Edoxaban70.4Rivaroxaban59.2
      3Dabigatran50.7Edoxaban62.3Dabigatran58.2Dabigatran48.6
      4Edoxaban26.5VKAs34.4Apixaban44.4Apixaban43
      5VKAs5.1Dabigatran5.8VKAs5.6VKAs16.3
      Prevention
      1Apixaban81.6Apixaban75.4Apixaban64.1Apixaban73.6
      2LMWH61.9LMWH51.4Edoxaban48.9LMWH59.9
      3Edoxaban57.3Dabigatran51.3LMWH48.7Edoxaban53.3
      4Dabigatran39.9Edoxaban48.9Dabigatran47.7Dabigatran43.3
      5Rivaroxaban9.4Rivaroxaban23.0Rivaroxaban40.4Rivaroxaban19.8
      VKA = vitamin K antagonist; LMWH = low molecular weight heparin.

      Gastrointestinal bleeding

      The NMA results for GI bleeding are shown in Figure 2B. Compared with apixaban, dabigatran (OR 3.48, 95% CI 1.41 – 8.60) and VKAs (OR 2.46, 95% CI 1.05 – 5.78) had a higher risk of bleeding, and the difference was significant. Compared with dabigatran, rivaroxaban (OR 0.40, 95% CI 0.17 – 0.95) and VKAs (OR 0.71, 95% CI 0.52 – 0.96) had a lower risk of bleeding, and the difference was significant.
      In the ranking of the cumulative probability of GI bleeding (Table 1 and Supplementary Fig. S2B), apixaban (SUCRA 80.7) had the highest safety, followed by rivaroxaban (SUCRA 66.8), edoxaban (SUCRA 62.3), VKAs (SUCRA 34.4), and dabigatran (SUCRA 5.8).

      Intracranial bleeding

      The NMA results for intracranial bleeding are shown in Figure 2C. Compared with VKAs, edoxaban (OR 3.61, 95% CI 1.34 – 9.73) and rivaroxaban (OR 4.03, 95% CI 1.13 – 14.29) had a lower risk of bleeding, and the difference was significant.
      In the ranking of the cumulative probability of intracranial bleeding (Table 1 and Supplementary Fig. S2C), rivaroxaban (SUCRA 74.4) was the safest, followed by edoxaban (SUCRA 70.4), dabigatran (SUCRA 58.2), apixaban (SUCRA 44.4), and VKAs (SUCRA 5.6).

      Fatal bleeding

      The NMA results for fatal bleeding are shown in Figure 2D. Edoxaban (OR 5.00, 95% CI 1.10 – 22.86) had a lower risk of bleeding than VKAs, and the difference was significant.
      In the ranking of the cumulative probability of fatal bleeding (Table 1 and Supplementary Fig. S2D), edoxaban (SUCRA 82.7) was the safest, followed by rivaroxaban (SUCRA 59.2), dabigatran (SUCRA 48.6), apixaban (SUCRA 43.0), and VKAs (SUCRA 16.3).

      Network meta-analysis results for the prevention of venous thromboembolism

      The NMA results for VTE prevention are shown in Figure 3. With regard to major bleeding, apixaban (OR 2.14, 95% CI 1.02 – 4.52) and LMWH (OR 0.58, 95% CI 0.36 – 0.95) had a lower risk of bleeding than rivaroxaban, and the difference was significant. In addition, there was no significant difference in the comparison of bleeding risk to other interventions.
      Figure 3
      Figure 3Forest plots of (A) major bleeding, (B) gastrointestinal bleeding, (C) intracranial bleeding, and (D) fatal bleeding related to the prevention of venous thromboembolism (VTE). Asterisk indicates statistical significance. CI = confidence interval; LMWH = low molecular weight heparin; Prl=prediction interval.
      In the ranking of the cumulative probability (Table 1 and Supplementary Fig. S3), for major bleeding, apixaban (SUCRA 81.6) was the safest, followed by LMWH (SUCRA 61.9), edoxaban (SUCRA 57.3), dabigatran (SUCRA 39.9), and rivaroxaban (SUCRA 9.4). For GI bleeding, apixaban (SUCRA 75.4) was the safest, followed by LMWH (SUCRA 51.4), dabigatran (SUCRA 51.3), edoxaban (SUCRA 48.9), and rivaroxaban (SUCRA 23.0). For intracranial bleeding, apixaban (SUCRA 64.1) was the safest, followed by edoxaban (SUCRA 48.9), LMWH (SUCRA 48.7), dabigatran (SUCRA 47.7), and rivaroxaban (SUCRA 40.4). Finally, for fatal bleeding, apixaban (SUCRA 73.6) was the safest, followed by LMWH (SUCRA 59.9), edoxaban (SUCRA 53.3), dabigatran (SUCRA 43.3), and rivaroxaban (SUCRA 19.8).

      Sensitivity analysis

      The sensitivity analysis results (see Supplementary Table S9) showed that for the treatment of VTE (lack of data on rivaroxaban in the acute VTE studies), the ranking of the two drugs (apixaban and dabigatran) changed but without any statistically significant difference in the risk of fatal bleeding (OR 1.04, 95% CI 0.06 – 19.49). For the prevention of VTE, the ranking of the two drugs (apixaban and rivaroxaban) changed but without any statistically significant difference in the risk of intracranial bleeding (OR 0.82, 95% CI 0.03 – 21.19); the rest of the rankings did not change. The sensitivity analysis demonstrated no significant change in the rank probability, which supported the robustness and reliability of the NMA.

      Publication bias

      Funnel charts and Egger test p values were generated to evaluate the publication bias of included studies (see Supplementary Fig. S4). The scattered points were basically distributed in a funnel symmetrical manner and the p value was > .05, which meant there was no significant publication bias.

      Discussion

      A NMA combines direct evidence (direct comparisons) and indirect evidence (comparison through other reference interventions). Combining direct and indirect evidence can improve the accuracy of estimation and provide estimates for all pairwise comparisons without direct evidence.
      • Caldwell D.M.
      An overview of conducting systematic reviews with network meta-analysis.
      The SUCRA was used to evaluate the advantages and disadvantages of intervention measures. A higher SUCRA value indicated a lower risk of bleeding and greater safety. This study is the first to use NMA to compare independently the risk of four severe types of bleeding (major, GI, intracranial, and fatal) that occurred as a result of DOACs used for the prevention and treatment of VTE.
      For major bleeding occurring during the treatment of VTE, in the NMA carried out by Mai et al.,
      • Mai V.
      • Bertoletti L.
      • Cucherat M.
      • Jardel S.
      • Grange C.
      • Provencher S.
      • et al.
      Extended anticoagulation for the secondary prevention of venous thromboembolic events: An updated network meta-analysis.
      apixaban had the lowest risk of bleeding and VKAs the highest. This finding is consistent with the present results. In the study of Sobieraj et al.,
      • Sobieraj D.M.
      • Coleman C.I.
      • Pasupuleti V.
      • Deshpande A.
      • Kaw R.
      • Hernandez A.V.
      Comparative efficacy and safety of anticoagulants and aspirin for extended treatment of venous thromboembolism: a network meta-analysis.
      the bleeding risk of dabigatran was lower than that of rivaroxaban, which is inconsistent with the present results. This inconsistency may be due to the combining of major bleeding and clinically relevant non-major bleeding in the study of Sobieraj et al.;
      • Sobieraj D.M.
      • Coleman C.I.
      • Pasupuleti V.
      • Deshpande A.
      • Kaw R.
      • Hernandez A.V.
      Comparative efficacy and safety of anticoagulants and aspirin for extended treatment of venous thromboembolism: a network meta-analysis.
      the studies included in the present study all used the ISTH definition of major bleeding.
      For GI bleeding occurring during the treatment of VTE, dabigatran had the highest risk, which may be related to its pharmacological properties. Dabigatran is administered as a pro-drug (dabigatran etexilate), and being activated through absorption in the GI mucosa, lead to an increased risk of GI bleeding.
      • Stangier J.
      • Clemens A.
      Pharmacology, pharmacokinetics, and pharmacodynamics of dabigatran etexilate, an oral direct thrombin inhibitor.
      In the NMA of Burr et al.,
      • Burr N.
      • Lummis K.
      • Sood R.
      • Kane J.S.
      • Corp A.
      • Subramanian V.
      Risk of gastrointestinal bleeding with direct oral anticoagulants: a systematic review and network meta-analysis.
      which compared VKAs and dabigatran, factor Xa inhibitors (apixaban, edoxaban, and rivaroxaban) reduced GI bleeding, and dabigatran was the drug with the highest bleeding risk, which is consistent with the present results. Although the data from the study of Burr et al. were highly consistent with the present results,
      • Burr N.
      • Lummis K.
      • Sood R.
      • Kane J.S.
      • Corp A.
      • Subramanian V.
      Risk of gastrointestinal bleeding with direct oral anticoagulants: a systematic review and network meta-analysis.
      a comparison of the differences in the risk of GI bleeding between different factor Xa inhibitors was not done. Guo et al. showed that the risk of GI bleeding of rivaroxaban was the highest;
      • Guo W.Q.
      • Chen X.H.
      • Tian X.Y.
      • Li L.
      Differences in gastrointestinal safety profiles among novel oral anticoagulants: evidence from a network meta-analysis.
      these data do not agree with the present results. The reason for this difference may be that the study of Guo et al. included six articles, with rivaroxaban as the experimental group,
      • Guo W.Q.
      • Chen X.H.
      • Tian X.Y.
      • Li L.
      Differences in gastrointestinal safety profiles among novel oral anticoagulants: evidence from a network meta-analysis.
      and included patients with VTE and atrial fibrillation (AF). Compared with VTE, the follow up duration for AF was longer, which may have increased the bleeding risk of rivaroxaban.
      For intracranial bleeding occurring during the treatment of VTE, the present NMA showed that rivaroxaban and edoxaban significantly reduced the risk of bleeding compared with VKAs. Moreover, the safety of the four DOACs was higher than that of VKAs, data which were consistent with the results of a conventional meta-analysis by Wu et al.
      • Wu T.
      • Lv C.
      • Wu L.
      • Chen W.
      • Lv M.
      • Jiang S.
      • et al.
      Risk of intracranial bleeding with direct oral anticoagulants: a systematic review and meta-analysis of randomized controlled trials.
      Therefore, combination of the present NMA and previous meta-analysis suggests that, compared with VKAs, DOACs may be safer against intracranial bleeding in patients with VTE. Compared with VKAs, the mechanism by which DOACs reduce the risk of intracranial bleeding is incompletely understood. This may be because DOACs target only one coagulation factor, whereas VKAs inhibit multiple coagulation factors.
      • Hartmann S.
      • Biliouris K.
      • Lesko L.J.
      • Nowak-Göttl U.
      • Trame M.N.
      Quantitative systems pharmacology model-based predictions of clinical endpoints to optimize warfarin and rivaroxaban anti-thrombosis therapy.
      DOACs seem to be safer overall, but the safety ranking of individual drugs differs. In the present study, rivaroxaban was the safest and dabigatran was less safety than rivaroxaban. However, another study showed that dabigatran was the safest.
      • Wolfe Z.
      • Khan S.U.
      • Nasir F.
      • Raghu Subramanian C.
      • Lash B.
      A systematic review and Bayesian network meta-analysis of risk of intracranial bleeding with direct oral anticoagulants.
      This observation was not found in the present study, but the studies mentioned above had one thing in common: they did not target patients with VTE and also included patients with AF, suggesting that different disease states may be associated with different bleeding risks. Therefore, an independent analysis of VTE and AF is important.
      For fatal bleeding occurring during the treatment of VTE, it was found that edoxaban significantly reduced the risk of bleeding compared with VKAs. In a conventional meta-analysis by Sharma et al.,
      • Sharma M.
      • Cornelius V.R.
      • Patel J.P.
      • Davies J.G.
      • Molokhia M.
      Efficacy and harms of direct oral anticoagulants in the elderly for stroke prevention in atrial fibrillation and secondary prevention of venous thromboembolism: systematic review and meta-analysis.
      the safety of the four DOACs was higher than that of VKAs, and edoxaban was the safest, data which are consistent with the present results. However, owing to the very small number of fatal bleeding events, there was no DOAC NMA related to fatal bleeding. Castellucci et al.
      • Castellucci L.A.
      • Cameron C.
      • Le Gal G.
      • Rodger M.A.
      • Coyle D.
      • Wells P.S.
      • et al.
      Efficacy and safety outcomes of oral anticoagulants and antiplatelet drugs in the secondary prevention of venous thromboembolism: systematic review and network meta-analysis.
      found that, although fatal bleeding was used as a secondary observation index, the small number of fatal events reported in the included studies led to inaccurate estimation of pooling points and, ultimately, no safety ranking of drugs.
      The different rankings of DOACs for different bleeding types may be caused by multiple factors. Currently, there is a lack of clarity on this issue. In some studies of VTE treatment, apixaban was the safest in terms of major bleeding and GI bleeding, probably because the dose of apixaban was 5 mg twice daily, which was the smallest single dose, compared with other drugs (see Appendix S1), and the smaller dose may be the reason for the high safety. However, with regard to intracranial bleeding, apixaban had the lowest safety, which may be due to the high binding rate of apixaban to the intracranial carrier protein, which made it easy to enter the blood–brain barrier (BBB). The high effective intracranial dose led to an increased risk of bleeding. The low binding rate of edoxaban with carrier protein makes it difficult to cross the BBB, which may be the reason for its high safety in intracranial bleeding.
      • Kakkos S.K.
      • Gohel M.
      • Baekgaard N.
      • Bauersachs R.
      • Bellmunt-Montoya S.
      • Black S.A.
      • et al.
      Editor's Choice – European Society for Vascular Surgery (ESVS) 2021 Clinical Practice Guidelines on the Management of Venous Thrombosis.
      After consulting the literature, it was found that most studies only rank bleeding risk and rarely explain the different physical and chemical factors that contribute to the ranking of the DOAC, which is worthy of further exploration.
      For the prevention of VTE, apixaban and rivaroxaban were the drugs with the highest and lowest bleeding risk among the four types of bleeding, respectively. This reminded us that in the prevention of VTE, the safety ranking of DOACs was relatively stable, and it was recommended that apixaban is used for the prevention of VTE. In the treatment of VTE, the safety ranking of anticoagulant drugs in the four bleeding types was less consistent than the results of prevention in VTE. This may be because the population in the prevention studies were essentially patients who had a hip or knee arthroplasty, or it may be because of the small drug doses in VTE prevention. Therefore, it is important to study separately the prevention and treatment studies. In the network meta-analysis of Hur et al.,
      • Hur M.
      • Park S.K.
      • Koo C.H.
      • Jung E.D.
      • Kang P.
      • Kim W.H.
      • et al.
      Comparative efficacy and safety of anticoagulants for prevention of venous thromboembolism after hip and knee arthroplasty.
      apixaban had the lowest risk of bleeding and rivaroxaban the highest. This was consistent with the present results and strengthened the recommendation of apixaban for the prevention of VTE.
      DOACs are available in a variety of doses, with different doses used for different indications. For example, rivaroxaban 10 mg once daily is used for the prevention of VTE after hip and knee arthroplasty, and 20 mg once daily for the treatment of VTE, while 15 mg once daily can also be used to treat AF. In the present study of the treatment of VTE, rivaroxaban had a high safety profile in intracranial bleeding. However, in the study of Wu et al.,
      • Wu T.
      • Lv C.
      • Wu L.
      • Chen W.
      • Lv M.
      • Jiang S.
      • et al.
      Risk of intracranial bleeding with direct oral anticoagulants: a systematic review and meta-analysis of randomized controlled trials.
      rivaroxaban had a low safety profile in intracranial bleeding, which is inconsistent with the present research results. It may be because in the current study only rivaroxaban 20 mg once a day for the treatment of VTE was included, while the indications in the study by Wu et al. included not only VTE, but also AF, so that rivaroxaban 15 mg once daily was also included. This shows that different indications have different doses, and different doses may lead to different research results.
      The clinical selection of an appropriate anticoagulant should not only consider the safety of the anticoagulant, but also the efficacy. However, the investigation of efficacy was not within the objectives of the current study.
      The present research had three main advantages. First, this is the first NMA of the four types of severe bleeding risk that occur as a result of DOACs used in the prevention and treatment of VTE. Second, the prevention and treatment trials of VTE were analysed separately, and the bleeding safety of the drugs ranked separately, which increased the reliability of the trial and reduced its heterogeneity. Third, different from previous NMAs, the prevention and treatment of VTE were studied as a single disease. Therefore, the data may be more targeted for patients with VTE in clinical practice.
      This research had four main limitations. First, the review protocol was not pre-specified or registered, which may cause issues of transparency, but the meta-analysis process followed the PRISMA guidelines strictly. Second, the NMA included 31 RCTs (seven studies on apixaban, eight studies on dabigatran, six studies on edoxaban, and 10 studies on rivaroxaban), but only indirect comparisons were made for DOACs. Third, the data did not permit an analysis of the effects of age or sex on the risks of bleeding or risks of comorbidities, such as reason for prophylaxis. Fourth, the duration of treatment in the 31 RCTs was not the same. Different durations of treatment may affect the incidence of bleeding and cause heterogeneity. Therefore, conducting real world research to confirm these findings would be worthwhile.

      Conclusion

      For the treatment of VTE, in terms of major and GI bleeding, apixaban had the lowest bleeding risk; in terms of intracranial bleeding, rivaroxaban had the lowest bleeding risk; and in terms of fatal bleeding, edoxaban had the lowest bleeding risk. For the prevention of VTE, apixaban had the lowest bleeding risk.

      Conflicts of interest

      None.

      Funding sources

      This study was funded by the Natural Science Foundation of Fujian Province , China ( 2018Y0037 ) and Fujian Provincial Health Technology Project , China ( 2019-CX-19 ).

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      References

        • Kakkos S.K.
        • Gohel M.
        • Baekgaard N.
        • Bauersachs R.
        • Bellmunt-Montoya S.
        • Black S.A.
        • et al.
        Editor's Choice – European Society for Vascular Surgery (ESVS) 2021 Clinical Practice Guidelines on the Management of Venous Thrombosis.
        Eur J Vasc Endovasc Surg. 2021; 61: 9-82
        • Heit J.A.
        • Spencer F.A.
        • White R.H.
        The epidemiology of venous thromboembolism.
        J Thromb Thrombolysis. 2016; 41: 3-14
        • Maynard G.
        Preventing Hospital-Associated Venous Thromboembolism: A Guide for Effective Quality Improvement.
        2nd edn. Agency for Healthcare Research and Quality, Rockville, MD2015 (AHRQ Publication No. 16-0001-EF)
        • Shermock K.M.
        • Lau B.D.
        • Haut E.R.
        • Hobson D.B.
        • Ganetsky V.S.
        • Kraus P.S.
        • et al.
        Patterns of non-administration of ordered doses of venous thromboembolism prophylaxis: implications for novel intervention strategies.
        PLoS One. 2013; 8e66311
        • Keeling D.
        • Baglin T.
        • Tait C.
        • Watson H.
        • Perry D.
        • Baglin C.
        • et al.
        Guidelines on oral anticoagulation with warfarin – fourth edition.
        Br J Haematol. 2011; 154: 311-324
        • Cohen A.T.
        • Berger S.E.
        • Milenković D.
        • Hill N.R.
        • Lister S.
        Anticoagulant selection for patients with VTE-Evidence from a systematic literature review of network meta-analyses.
        Pharmacol Res. 2019; 143: 166-177
        • Chaudhary R.
        • Sharma T.
        • Garg J.
        • Sukhi A.
        • Bliden K.
        • Tantry U.
        • et al.
        Direct oral anticoagulants: a review on the current role and scope of reversal agents.
        J Thromb Thrombolysis. 2020; 49: 271-286
        • Harter K.
        • Levine M.
        • Henderson S.O.
        Anticoagulation drug therapy: a review.
        West J Emerg Med. 2015; 16: 11-17
        • Gomez-Outes A.
        • Terleira-Fernandez A.I.
        • Suarez-Gea M.L.
        • Vargas-Castrillon E.
        Dabigatran, rivaroxaban, or apixaban versus enoxaparin for thromboprophylaxis after total hip or knee replacement: systematic review, meta-analysis, and indirect treatment comparisons.
        BMJ. 2012; 344: e3675
        • Büller H.R.
        • Prins M.H.
        • Lensin A.W.
        • Decousus H.
        • Jacobson B.F.
        • et al.
        • EINSTEIN–PE Investigators
        Oral rivaroxaban for the treatment of symptomatic pulmonary embolism.
        N Engl J Med. 2012; 366: 1287-1297
        • Cohen A.T.
        • Spiro T.E.
        • Büller H.R.
        • Haskell L.
        • Hu D.
        • Hull R.
        • et al.
        Rivaroxaban for thromboprophylaxis in acutely ill medical patients.
        N Engl J Med. 2013; 368: 513-523
        • Hutton B.
        • Salanti G.
        • Caldwell D.M.
        • Chaimani A.
        • Schmid C.H.
        • Cameron C.
        • et al.
        The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations.
        Ann Intern Med. 2015; 162: 777-784
        • Schulman S.
        • Kearon C.
        Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients.
        J Thromb Haemost. 2005; 3: 692-694
        • Sterne J.A.C.
        • Savović J.
        • Page M.J.
        • Elbers R.G.
        • Blencowe N.S.
        • Boutron I.
        • et al.
        RoB 2: a revised tool for assessing risk of bias in randomised trials.
        BMJ. 2019; 366: l4898
        • Salanti G.
        • Del Giovane C.
        • Chaimani A.
        • Caldwell D.M.
        • Higgins J.P.
        Evaluating the quality of evidence from a network meta-analysis.
        PLoS One. 2014; 9e99682
        • Agnelli G.
        • Buller H.R.
        • Cohen A.
        • Curto M.
        • Gallus A.S.
        • Johnson M.
        • et al.
        Oral apixaban for the treatment of acute venous thromboembolism.
        N Engl J Med. 2013; 369: 799-808
        • Buller H.
        • Deitchman D.
        • Prins M.
        • Segers A.
        • Botticelli Investigators, Writing Committe
        Efficacy and safety of the oral direct factor Xa inhibitor apixaban for symptomatic deep vein thrombosis. The Botticelli DVT dose-ranging study.
        J Thromb Haemost. 2008; 6: 1313-1318
        • Buller H.R.
        • Lensing A.W.
        • Prins M.H.
        • Agnelli G.
        • Cohen A.
        • Gallus A.S.
        • et al.
        A dose-ranging study evaluating once-daily oral administration of the factor Xa inhibitor rivaroxaban in the treatment of patients with acute symptomatic deep vein thrombosis: the Einstein-DVT Dose-Ranging Study.
        Blood. 2008; 112: 2242-2247
        • Büller H.R.
        • Décousus H.
        • Grosso M.A.
        • Mercuri M.
        • Middeldorp S.
        • et al.
        • Hokusai-VTE Investigators
        Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism.
        N Engl J Med. 2013; 369: 1406-1415
        • Bauersachs R.
        • Berkowitz S.D.
        • Brenner B.
        • Buller H.R.
        • Decousus H.
        • et al.
        • EINSTEIN Investigators
        Oral rivaroxaban for symptomatic venous thromboembolism.
        N Engl J Med. 2010; 363: 2499-2510
        • Eriksson B.I.
        • Dahl O.E.
        • Büller H.R.
        • Hettiarachchi R.
        • Rosencher N.
        • Bravo M.L.
        • et al.
        A new oral direct thrombin inhibitor, dabigatran etexilate, compared with enoxaparin for prevention of thromboembolic events following total hip or knee replacement: the BISTRO II randomized trial.
        J Thromb Haemost. 2005; 3: 103-111
        • Eriksson B.I.
        • Borris L.C.
        • Dahl O.E.
        • Haas S.
        • Huisman M.V.
        • Kakkar A.K.
        • et al.
        A once-daily, oral, direct Factor Xa inhibitor, rivaroxaban (BAY 59-7939), for thromboprophylaxis after total hip replacement.
        Circulation. 2006; 114: 2374-2381
        • Eriksson B.I.
        • Dahl O.E.
        • Rosencher N.
        • Kurth A.A.
        • van Dijk C.N.
        • Frostick S.P.
        • et al.
        Dabigatran etexilate versus enoxaparin for prevention of venous thromboembolism after total hip replacement: a randomised, double-blind, non-inferiority trial.
        Lancet. 2007; 370: 949-956
        • Eriksson B.I.
        • Dahl O.E.
        • Rosencher N.
        • Kurth A.A.
        • van Dijk C.N.
        • Frostick S.P.
        • et al.
        Oral dabigatran etexilate vs. subcutaneous enoxaparin for the prevention of venous thromboembolism after total knee replacement: the RE-MODEL randomized trial.
        J Thromb Haemost. 2007; 5: 2178-2185
        • Eriksson B.I.
        • Borris L.C.
        • Friedman R.J.
        • Haas S.
        • Huisman M.V.
        • Kakkar A.K.
        • et al.
        Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty.
        N Engl J Med. 2008; 358: 2765-2775
        • Eriksson B.I.
        • Dahl O.E.
        • Huo M.H.
        • Kurth A.A.
        • Hantel S.
        • Hermansson K.
        • et al.
        Oral dabigatran versus enoxaparin for thromboprophylaxis after primary total hip arthroplasty (RE-NOVATE II∗). A randomised, double-blind, non-inferiority trial.
        Thromb Haemost. 2011; 105: 721-729
        • Fuji T.
        • Wang C.J.
        • Fujita S.
        • Kawai Y.
        • Kimura T.
        • Tachibana S.
        Safety and efficacy of edoxaban, an oral factor xa inhibitor, for thromboprophylaxis after total hip arthroplasty in Japan and Taiwan.
        J Arthroplasty. 2014; 29: 2439-2446
        • Fuji T.
        • Wang C.J.
        • Fujita S.
        • Kawai Y.
        • Nakamura M.
        • Kimura T.
        • et al.
        Safety and efficacy of edoxaban, an oral factor Xa inhibitor, versus enoxaparin for thromboprophylaxis after total knee arthroplasty: the STARS E-3 trial.
        Thromb Res. 2014; 134: 1198-1204
        • Fuji T.
        • Fujita S.
        • Kawai Y.
        • Nakamura M.
        • Kimura T.
        • Kiuchi Y.
        • et al.
        Safety and efficacy of edoxaban in patients undergoing hip fracture surgery.
        Thromb Res. 2014; 133: 1016-1022
        • Fuji T.
        • Fujita S.
        • Kawai Y.
        • Nakamura M.
        • Kimura T.
        • Fukuzawa M.
        • et al.
        Efficacy and safety of edoxaban versus enoxaparin for the prevention of venous thromboembolism following total hip arthroplasty: STARS J-V.
        Thromb J. 2015; 13: 27
        • Ginsberg J.S.
        • Davidson B.L.
        • Comp P.C.
        • Francis C.W.
        • Friedman R.J.
        • et al.
        • RE-MOBILIZE Writing Committee
        Oral thrombin inhibitor dabigatran etexilate vs North American enoxaparin regimen for prevention of venous thromboembolism after knee arthroplasty surgery.
        J Arthroplasty. 2009; 24: 1-9
        • Kakkar A.K.
        • Brenner B.
        • Dahl O.E.
        • Eriksson B.I.
        • Mouret P.
        • Muntz J.
        • et al.
        Extended duration rivaroxaban versus short-term enoxaparin for the prevention of venous thromboembolism after total hip arthroplasty: a double-blind, randomised controlled trial.
        Lancet. 2008; 372: 31-39
        • Lassen M.R.
        • Davidson B.L.
        • Gallus A.
        • Pineo G.
        • Ansell J.
        • Deitchman D.
        The efficacy and safety of apixaban, an oral, direct factor Xa inhibitor, as thromboprophylaxis in patients following total knee replacement.
        J Thromb Haemost. 2007; 5: 2368-2375
        • Lassen M.R.
        • Ageno W.
        • Borris L.C.
        • Lieberman J.R.
        • Rosencher N.
        • Bandel T.J.
        • et al.
        Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty.
        N Engl J Med. 2008; 358: 2776-2786
        • Lassen M.R.
        • Raskob G.E.
        • Gallus A.
        • Pineo G.
        • Chen D.
        • Portman R.J.
        Apixaban or enoxaparin for thromboprophylaxis after knee replacement.
        N Engl J Med. 2009; 361: 594-604
        • Lassen M.R.
        • Raskob G.E.
        • Gallus A.
        • Pineo G.
        • Chen D.
        • Hornick P.
        • et al.
        Apixaban versus enoxaparin for thromboprophylaxis after knee replacement (ADVANCE-2): a randomised double-blind trial.
        Lancet. 2010; 375: 807-815
        • Lassen M.R.
        • Raskob G.E.
        • Gallus A.
        • Pineo G.
        • Chen D.
        • Hornick P.
        • et al.
        Apixaban versus enoxaparin for thromboprophylaxis after hip replacement.
        N Engl J Med. 2010; 363: 2487-2498
        • Nakamura M.
        • Nishikawa M.
        • Komuro I.
        • Kitajima I.
        • Uetsuka Y.
        • Yamagami T.
        • et al.
        Apixaban for the treatment of japanese subjects with acute venous thromboembolism (AMPLIFY-J Study).
        Circ J. 2015; 79: 1230-1236
        • Raskob G.
        • Cohen A.T.
        • Eriksson B.I.
        • Puskas D.
        • Shi M.
        • Bocanegra T.
        • et al.
        Oral direct factor Xa inhibition with edoxaban for thromboprophylaxis after elective total hip replacement. A randomised double-blind dose-response study.
        Thromb Haemost. 2010; 104: 642-649
        • Schulman S.
        • Kearon C.
        • Kakkar A.K.
        • Mismetti P.
        • Schellong S.
        • Eriksson H.
        • et al.
        Dabigatran versus warfarin in the treatment of acute venous thromboembolism.
        N Engl J Med. 2009; 361: 2342-2352
        • Schulman S.
        • Kakkar A.K.
        • Goldhaber S.Z.
        • Schellong S.
        • Eriksson H.
        • Mismetti P.
        • et al.
        Treatment of acute venous thromboembolism with dabigatran or warfarin and pooled analysis.
        Circulation. 2014; 129: 764-772
        • Schulman S.
        • Kearon C.
        • Kakkar A.K.
        • Schellong S.
        • Eriksson H.
        • Baanstra D.
        • et al.
        Extended use of dabigatran, warfarin, or placebo in venous thromboembolism.
        N Engl J Med. 2013; 368: 709-718
        • Samama C.M.
        • Laporte S.
        • Rosencher N.
        • Girard P.
        • Llau J.
        • Mouret P.
        • et al.
        Rivaroxaban or enoxaparin in nonmajor orthopedic surgery.
        N Engl J Med. 2020; 382: 1916-1925
        • Turpie A.G.
        • Lassen M.R.
        • Davidson B.L.
        • Bauer K.A.
        • Gent M.
        • Kwong L.M.
        • et al.
        Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty (RECORD4): a randomised trial.
        Lancet. 2009; 373: 1673-1680
        • Caldwell D.M.
        An overview of conducting systematic reviews with network meta-analysis.
        Syst Rev. 2014; 3: 109
        • Mai V.
        • Bertoletti L.
        • Cucherat M.
        • Jardel S.
        • Grange C.
        • Provencher S.
        • et al.
        Extended anticoagulation for the secondary prevention of venous thromboembolic events: An updated network meta-analysis.
        PLoS One. 2019; 14e0214134
        • Sobieraj D.M.
        • Coleman C.I.
        • Pasupuleti V.
        • Deshpande A.
        • Kaw R.
        • Hernandez A.V.
        Comparative efficacy and safety of anticoagulants and aspirin for extended treatment of venous thromboembolism: a network meta-analysis.
        Thromb Res. 2015; 135: 888-896
        • Stangier J.
        • Clemens A.
        Pharmacology, pharmacokinetics, and pharmacodynamics of dabigatran etexilate, an oral direct thrombin inhibitor.
        Clin Appl Thromb Hemost. 2009; 15: 9S-16S
        • Burr N.
        • Lummis K.
        • Sood R.
        • Kane J.S.
        • Corp A.
        • Subramanian V.
        Risk of gastrointestinal bleeding with direct oral anticoagulants: a systematic review and network meta-analysis.
        Lancet Gastroenterol Hepatol. 2017; 2: 85-93
        • Guo W.Q.
        • Chen X.H.
        • Tian X.Y.
        • Li L.
        Differences in gastrointestinal safety profiles among novel oral anticoagulants: evidence from a network meta-analysis.
        Clin Epidemiol. 2019; 11: 911-921
        • Wu T.
        • Lv C.
        • Wu L.
        • Chen W.
        • Lv M.
        • Jiang S.
        • et al.
        Risk of intracranial bleeding with direct oral anticoagulants: a systematic review and meta-analysis of randomized controlled trials.
        J Neurol. 2021; ([Epub ahead of print])https://doi.org/10.1007/s00415-021-10448-2
        • Hartmann S.
        • Biliouris K.
        • Lesko L.J.
        • Nowak-Göttl U.
        • Trame M.N.
        Quantitative systems pharmacology model-based predictions of clinical endpoints to optimize warfarin and rivaroxaban anti-thrombosis therapy.
        Front Pharmacol. 2020; 11: 1041
        • Wolfe Z.
        • Khan S.U.
        • Nasir F.
        • Raghu Subramanian C.
        • Lash B.
        A systematic review and Bayesian network meta-analysis of risk of intracranial bleeding with direct oral anticoagulants.
        J Thromb Haemost. 2018; 16: 1296-1306
        • Sharma M.
        • Cornelius V.R.
        • Patel J.P.
        • Davies J.G.
        • Molokhia M.
        Efficacy and harms of direct oral anticoagulants in the elderly for stroke prevention in atrial fibrillation and secondary prevention of venous thromboembolism: systematic review and meta-analysis.
        Circulation. 2015; 132: 194-204
        • Castellucci L.A.
        • Cameron C.
        • Le Gal G.
        • Rodger M.A.
        • Coyle D.
        • Wells P.S.
        • et al.
        Efficacy and safety outcomes of oral anticoagulants and antiplatelet drugs in the secondary prevention of venous thromboembolism: systematic review and network meta-analysis.
        BMJ. 2013; 347: f5133
        • Hur M.
        • Park S.K.
        • Koo C.H.
        • Jung E.D.
        • Kang P.
        • Kim W.H.
        • et al.
        Comparative efficacy and safety of anticoagulants for prevention of venous thromboembolism after hip and knee arthroplasty.
        Acta Orthop. 2017; 88: 634-641

      Comments

      Commenting Guidelines

      To submit a comment for a journal article, please use the space above and note the following:

      • We will review submitted comments as soon as possible, striving for within two business days.
      • This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
      • We require that commenters identify themselves with names and affiliations.
      • Comments must be in compliance with our Terms & Conditions.
      • Comments are not peer-reviewed.